Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Happy Birthday AP 2.0

This site may earn commission on affiliate links.
IMO, tesla is only obligated to upgrade those that exercised their FSD option. I believe Tesla has consistently operated that if an option is not secured before it lapses (or is revoked), then the customer is SOL.

There is a contract law basis for this since it isn't an "option contract" but rather an offering which may be accepted by a customer (but offers are only around as long as they are conditioned to be open or until they are revoked by either the offering party or the offered party).

Thus, I expect Tesla to only upgrade those that actually bought FSD (aka us suckers) and not those that didn't (even though they advertised the car has having the hardware for FSD, I believe there is a standing issue that precludes or stands as a serious hurdle to enforcing any perceived damages from a lack of upgrade). It appears the HW2 will be sufficient for EAP (though that remains to be seen).

So I don't think everyone will be getting Xavier (or anything). We'll have to wait and see how it shakes out.
BS, they said they were selling us cars with FSD hardware installed. Just because you got suckered to pay up front doesn’t mean they don’t have an obligation to the rest of us.
 
BS, they said they were selling us cars with FSD hardware installed. Just because you got suckered to pay up front doesn’t mean they don’t have an obligation to the rest of us.

I'm not saying you can't make an argument that you wouldn't have bought your particular car if it weren't fully enabled and ready for FSD but I'm just saying don't expect the $4k option to remain $4k forever or even be around. Tesla has, and likely will continue to, change the terms of these offers and if you haven't accepted the offer, it isn't guaranteed to stick around.

So if you really want the FSD, I'd buy it otherwise it is a risk.

But I'm a sucker, so I wouldn't take my advice on giving Tesla more money. Even now we're still only getting close to AP1 parity.

Yours, Electroman, is a unique situation where you have HW2 but paid for AP1. So I don't even know what to make of that. I also wonder where you'll end up on the software side -- it seems like a good possibility that you get EAP without paying for it. I can't imagine its worth the effort to fork the code or hide car options but maybe Tesla will do that. I'm not a software whiz.
 
Works for me better than 50% but not close enough to 100% to make it useful. My garage door is fairly narrow -- I've got about 7-8 inches to spare (3-4" on either side) with the windows folded in. I also have a retaining wall on either side of the narrow driveway, and the garage itself is narrow (1-car). If it's not pretty well lined up to begin with, there's no chance. But even if it is lined up nice and straight, some significant fraction of the time it will freak out and turn itself crooked for no good reason. Once it's even slightly crooked it cannot recover. (I'd estimate this happens 10-20% of the time. I think the feature starts to get useful if it fails <5% of the time; any more than that and it's not worth standing around with your neighbors staring at you fiddling with the thing.)

The problem is that using the ultrasonics it has, this is a very hard problem. It has blind spots directly to the sides, and not enough resolution to clearly perceive the opening, particularly after it's started to go trough the door and the garage door sill is between two sensors. (Remember, ultrasonic sensors give you just a range, not a direction. You don't get a coordinate in 3D space; you get a semi-spherical surface centered on the sensor.) I think that maybe a very sophisticated software approach could work here with only the ultrasonic sensors; it would need to build itself a 3D map based on building up ultrasonic returns over time as it proceeds and (roughly speaking) taking the intersection of these spheres (but not that simple as there is always uncertainty). As it is currently, it seems to forget that the door sill is there once it's passed the sensors.

I have some hope that they will eventually start using the cameras to help with summon. They could use a structure-from-motion algorithm to build up a 3D map of the environment. If they remember this map not just for the current parking maneuver but also remember it for next time so it's not starting from scratch, they might be able to do something reasonable. (In particular, if you've just parked in a tight garage, you'd want to remember the map you built up when you parked to have something to start with -- you only get structure from motion after you start moving...)

I see absolutely zero evidence that they have even begun to implement such a sophisticated approach -- despite the fact that what I just described is not some fantastic new innovation requiring breakthroughs to implement. Just google "structure from motion" and you'll find tutorials and pre-built libraries[1]. They have plenty of compute power to do this. Tesla is just so slammed trying to get basic functionality working (and generally having too many balls in the air for such a relatively small company) that they are leaving low-hanging fruit like this to rot on the vine. Instead as has already been pointed out, they basically spent the past year trying to reproduce the MobileEye system with their own drop-in replacement. Elon obviously expected that to take about 2 months and then they'd be on to a new architecture with features like I just described that would not be feasible with the ME system.

[1] LMGTFY: Structure from motion - Wikipedia
Okay I must admit you have a significantly tighter garage fit than I do.
 
What's important to me is whether Elon lied to us about AP2, or if he was just horribly wrong. Autopilot program Director Sterling Anderson left in December to start his own self-driving startup. Maybe Sterling mislead Elon about what AP2 was capable of. It just seems really strange that Elon would flat out lie about it.

I think it was a result of a mixture of hubris, survival instincts, and the unwillingness to partner up.

Hubris because he had been stacking up win after win.
Survival instincts because Tesla has been running pedal to the metal for awhile now with no room to take a breather.
Unwillingness to partner up due to the what happened with MobileEye.

I also think he lost touch with the Tesla consumer that was becoming less and less hardcore fans, and more regular folks. Folks that are way less lenient.

I do believe AP2 in itself was a necessary lie when it came to the timeframe. But, I think most people could have forgiven him for that.

The problem was FSD. All the allowance for exaggeration was already taken up with AP2/EAP.

He essentially gambled when he was already winning the hearts/minds of people. It was a risk not worth taking, but he took it.

Regardless of why I think it's time to move on.

Right now the most exciting place to be on TMC or really the internet entirely is the HW 2.5 capabilities forum.

Why is it exciting? Because AP2/AP2.5 has hardware on it that can run really sophisticated deep neural nets, and a person has rooted his car to get access to it. Then someone else who has more expertise at neural nets has been trying to decipher what it does. So it's not just a bunch of people speculating based on no information. It's people looking at code, and making educated guesses.

As far as I'm concerned the bad times are over, and the fun is starting to begin. :)
 
I do believe AP2 in itself was a necessary lie when it came to the timeframe. But, I think most people could have forgiven him for that.

Good analysis there for the most part @S4WRXTTCS.

While I don't agree a lie was necessary or acceptable (in another thread I've been making the point they could have just released the hardware and announced software only once released), I think even if we assume AP was a necessary lie, EAP was not. Tesla could have done it the same way that AEB came to HW2.5, announce merely parity coming - not a superior feature.

Think of how different our conversations would be had Tesla in October 2016 merely announced upcoming AP1 parity for AP2. Nothing more promised, nothing "enhanced", no more money taken etc... The Design Studio would be selling the Autopilot 1 still today (on AP2 HW) think about the difference. There could be a small disclaimer saying the car has more cameras that are not used at this time.

As for letting it go, personally to an extent I agree. Not because I disagree with people who do not want to let go - that is their right until Tesla delivers and even then the misled past remains a stain on Elon/Tesla. It remains a valid topic for a long time. But I agree personally, it is a tiresome debate because it gets nowhere amongst people who on both sides have made up their minds.

At least the technical conversation keeps evolving, you know... Back in the Model X waiting days I was infamous for analyzing Tesla muleshots for details of the cars. That's how we found out about the adaptive spoiler months in advance. The HW2.5 thread reminds me of similar analysis, and even better we now have the product to explore. I don't have much to contribute to that myself from a technical ability point, but I agree it is a great thread (and the AP2 capabilities thread before it where we e.g. mapped the camera blind-spots and found out about Intel in Model 3 months in advance) to keep up to date with.

Keep up the great work @verygreen, @lunitiks and others!

Here is my suggestion in the other thread how Tesla could avoid overpromise-underdeliver without changing anything in their hardware or software update pace:

They are, but see how it has turned over time for them. To keep the demand levers swinging, they have time and time again resorted to selling current cars based on future features what would come to those cars... and now by AP2 at least, also over-promising and under-delivering on such features.

It doesn't automatically come with the territory. When AP2 came out, Tesla could still have released the hardware, could still have offered all the same future software updates, but they didn't need to paint the arguably misleading picture that they did. What if AP2 has been launched something like this:

"From today, all Tesla's ship with Autopilot 2 hardware: 8 cameras, 1 radar, 12 ultrasonics and an Nvidia supercomputer. At the first stage the car will not have any features using this hardware, though, so it is free with the car, but once more features are available, we'll let you know of the price of those features."

Same with AP1, they did not need to announce that it will one day read traffic lights and meet/find you at the curb and go from highway to highway and whatnot. They could have just said they are now shipping this hardware, period. Announce the software when it is shipping. Ask money for major software features when they are shipping...

That would have been selling what they have now and it would be a perfect approach IMO. It is a slippery slope once you start piling promise on promise and then need to promise even more to keep the attention...
 
Last edited:
  • Love
Reactions: FlatSix911
Yeah, I too think Tesla would probably be on a shaky ground by taking away the option to upgrade to FSD - especially if they do before people actually have even had a chance to upgrade to a released FSD. Taking it away one day, perhaps with warning ahead, is one thing. Taking it away before FSD is even released seems lot more iffier...

However I could totally see them raising the price a lot, if need be.
 
In all seriousness how are you liking .40 /.42? Happy with the driving experience now? Improved for you?
Improved straight-line and smoother curve at highway speeds, but have limited miles on .42 so far. That one run though has been the best to date.

My true test comes later today on a stretch that has diagonal tar lines that gave the car fits. Will have to try earlier to approximate sun position as it was about 8 weeks ago and Tesla questioned sun glare in their “we’ll pull the logs” delayed response. Doubt it was a factor as the road segment was S/E late in the day. Still waiting to hear back, but there was a 17.34 install before they responded, and it’s not clear whether the logs are cleared or not...
 
I think they had to lie. If MobilEye broke the contract (or did not renew) cause they did not want supply chips, what would happen to sales in 2017after that if they sold cars inferior to old ones? And all the AP1 backorders, customers would cancel and sue. Damned if you do, damned if you don't.